CN105473717B

CN105473717B - Antisense oligonucleotide composition

Info

Publication number: CN105473717B
Application number: CN201480046298.8A
Authority: CN
Inventors: S·C·W·理查森; P·D·R·戴尔; J·C·米歇尔
Original assignee: University of Greenwich
Current assignee: University of Greenwich
Priority date: 2013-06-21
Filing date: 2014-06-23
Publication date: 2020-05-01
Anticipated expiration: 2034-06-23
Also published as: JP6506270B2; ES2714531T3; WO2014203008A1; HK1223653A1; WO2014203008A9; EP3011025B1; US10400241B2; KR20160023784A; AU2014282986A1; CN105473717A; CA2923849A1; US20180251768A1; IL243174B; US9902960B2; AU2014282986B2; KR102266092B1; IL243174A0; US20160152984A1; JP2016526541A; CA2923849C

Abstract

The present invention relates to antisense oligonucleotide (ASO) compositions, and in particular to compositions and methods for cytoplasmic delivery of antisense oligonucleotides (ASO). Hybrid ASOs are provided that are partially single-stranded and partially double-stranded, which hybridize to form a double-stranded region that can be non-covalently bonded to a region of a nucleic acid binding protein. In this way, ASO:: protein complexes can be produced that facilitate the delivery of antisense DNA into target cells. Such complexes can be used to down-regulate gene expression in a cell.

Description

Antisense oligonucleotide composition

Technical Field

The present invention relates to antisense oligonucleotide (ASO) compositions, and in particular to compositions and methods for antisense oligonucleotide (ASO) cytoplasmic delivery. Hybrid ASOs are provided that are partially single-stranded and partially double-stranded, and hybridize to form a double-stranded region that can be non-covalently bonded to a nucleic-acid-binding protein regions. In this way, ASO:: protein complexes can be generated that facilitate the delivery of antisense DNA into target cells. Such complexes can be used to down-regulate gene expression in a cell.

Background

Antisense oligonucleotides have been approved by the FDA for use as antiviral agents in the treatment of cytomegalovirus (cytomegalovirus) -mediated retinitis and chronic ulcerative colitis (Roehr, 1998; Yacychyn et al, 1998). ASOs include single-stranded DNA fragments or analogs thereof designed to hybridize to messenger rna (mrna) transcripts derived from a particular gene. The mRNA thus formed was ASO hybrids degraded by RNAse H. In virus-infected organisms, the viral genetic material enters specific cells to replicate, a process that requires translation of viral-specific mRNA. By treating infected cells with antisense oligonucleotides specific for viral mRNA, target viral gene expression can be prevented and the viral life cycle blocked.

An important factor in the effectiveness of antisense oligonucleotide therapy is the bioavailability of the oligonucleotide (Biroccio et al, 2003). Bioavailability may be limited by the inability of antisense oligonucleotides to penetrate the plasma membrane or endomembrane system of a cell. Since the target (e.g., mRNA) of antisense oligonucleotides is located within the cytosol of a cell, the oligonucleotides need to be able to cross the cell membrane before they approach the cytosol.

One way to address the problem of cytoplasmic access (cytosolic access) of such antisense oligonucleotides is to administer the treatment locally rather than systemically, in particular to the independent pharmacokinetic compartment, increasing the local concentration of antisense oligonucleotides. For example, antisense oligonucleotide therapy of cytomegalovirus-mediated retinitis involves intravitreal injection of fomivirsen (fomivirsen)

Administered to an independent pharmacokinetic compartment. However, cytoplasmic access, i.e., access of antisense oligonucleotides to their target mRNA, remains a significant limitation even after intravitreal administration (Lysik and Wu-Pong, 2003).

Other methods of intracellular delivery of antisense oligonucleotides may involve: mechanical and electrical cell damage to cell membranes, macromolecular and lipid carriers (carriers) known to non-specifically destabilize membranes, or viral vectors. However, these and similar methods have not proven clinically safe or reliable, and (to date) no approved delivery technology for antisense products has been generated on the market.

Disclosure of Invention

The present invention aims to provide compositions and methods for use in a system for down-regulating one or more genes within a cell by means of a novel arrangement of an active antisense sequence (e.g. of single-stranded DNA) flanking a partially overlapping second oligonucleotide strand forming a binding site for a nucleic acid binding domain that can bind to a protein. We named these partially single-stranded and partially double-stranded nucleotides "ASO hybrids". Unexpectedly, we have found and have shown empirically (unpublished data):

1) ASO hybrids can show a profile of antisense activity comparable to control single stranded antisense oligonucleotides in cell-free assays, e.g., at ASO concentrations of 30 pMol.

2) ASO hybrids bind to nucleic acid binding domains (e.g., saccharomyces cerevisiae (s. cerevisiae) GAL4) to form complexes without the need for polycationic affinity handles (e.g., poly (L-lysine) (Gaur et al, 2002; WO97/23236)) or other covalent chemical conjugation methods previously used.

3) If in an ASO hybridization complex (e.g., GAL4:: ASO), the nucleic acid binding domain is fused to attenuated lethal factor domain 1 (from Bacillus anthracis, i.e., LFn), the supramolecular assembly can pass through a pore derived from a bacterial virulence factor (Bacillus anthracis PA 63). This is unexpected, as it has been reported that passage through the PA pore requires structural disassembly (molten globular transition) of cargo (cargo) (Zornetta et al, 2010). For the conversion of the supramolecular complex into the cytosol via the PA pore, a change in the globular state of the cargo, i.e. a change from an ordered structure to a random coil, must not occur. Again, this conversion has been shown without the use of polycationic affinity treatments such as poly (L-lysine) (Gaur et al, 2002: WO 97/23236).

It is a particular object of the invention to provide a conjugation strategy that enables a composition of antisense oligonucleotides to cross the cell membrane. In this way, antisense oligonucleotides can be delivered to the cytosol of a cell, with the described antisense-to-protein conjugation strategy providing direct access to RNA or DNA targets (e.g., viral mRNA transcripts) within the cytosol.

In one aspect, the invention provides antisense oligonucleotide compositions comprising a pair of antisense oligonucleotides, wherein the two antisense oligonucleotides hybridize to give at least one single-stranded antisense sequence and at least one double-stranded protein-binding sequence. Preferably, the composition further comprises a shuttle protein (e.g., recombinant LFn-GAL4), wherein the shuttle protein comprises a nucleic acid binding domain (e.g., GAL4) that recognizes the double-stranded protein binding sequence, and wherein the shuttle protein is non-covalently bonded to the pair of antisense oligonucleotides.

In another aspect, the invention provides a system for delivering an antisense oligonucleotide across the membrane of a cell, the system comprising (i) a pair of antisense oligonucleotides, the pair comprising a double-stranded protein binding sequence (e.g., GAL4) and at least one single-stranded antisense sequence (e.g., as shown in FIG. 1), and (ii) a shuttle protein having a sequence (e.g., CGG-N) that recognizes a specific sequence contained within the double-stranded protein binding nucleic acid sequence₁₁-CCG) such that the shuttle protein non-covalently binds to the pair of antisense oligonucleotides.

In another aspect, the invention provides a method of non-covalently conjugating an antisense oligonucleotide to a shuttle protein (e.g., LFn-GAL4), the method comprising: (i) providing two antisense oligonucleotides, (ii) hybridizing the two antisense oligonucleotides to form an antisense hybrid, wherein the antisense hybrid comprises at least one single-stranded antisense sequence and at least one double-stranded protein-binding sequence, (iii) providing a shuttle protein comprising a nucleic acid-binding domain that recognizes the double-stranded protein-binding target sequence, and (iv) non-covalently conjugating the shuttle protein to the antisense hybrid via the nucleic acid-binding domain of the protein and the protein-binding sequence of the antisense hybrid.

The methods of conjugation and membrane translocation (and see table 1 below) do not require the use of (polycationic or ionic) DNA condensing agents (such as poly (ethyleneimine) or poly (L-lysine)) because these materials are generally toxic and this toxicity is believed to be due, at least in part, to their propensity to destabilize the cell membrane (Richardson et al, 1999) (see also table 1 below). In the context of the present invention, a shuttle protein is defined as a protein capable of transporting genetic material through the pores of a cell membrane. The pore may comprise a pore-forming protein, such as bacillus anthracis PA83 or PA 63.

The oligonucleotides of the invention include nucleic acid sequences which may be of DNA or DNA analogues (e.g. sulphurised analogues in which some of the oxygen atoms are replaced by sulphur). Other DNA analogs that may find suitable include those disclosed by Leumann (2002). The sequences may be directly adjacent to each other or linked by intervening DNA sequences, preferably at least partially double stranded.

In a preferred embodiment, the shuttle protein is an attenuated toxin protein. In particular, the shuttle protein may be the lethal factor domain I (LFn) of bacillus anthracis.

Preferably, in the attenuated toxin, at least one toxin domain, such as one or more of toxin domains II-IV of the Bacillus anthracis lethal factor protein toxin, is replaced with a nucleic acid binding domain. For example, the nucleic acid binding domain can be Saccharomyces cerevisiae (Saccharomyces cerevisiae) GAL4 (fused to LFn).

Advantageously, the antisense oligonucleotide composition further comprises a pore-forming protein. Preferably, the pore-forming protein is a non-toxic protein. One preferred pore-forming protein is the Protective Antigen of the virulence factor of Bacillus anthracis (PA). In particular, the pore-forming protein may be bacillus anthracis PA83 or PA 63.

In one embodiment (in the example where synaxin 5(syntaxin5, Syn5) is the target for down-regulation), one of the oligonucleotides includes the antisense sequence 5 'AATTTGTTTGTTGAGGCTA 3' (SEQ ID No: 18). Each member of the antisense oligonucleotide pair comprises one of the two complementary strands that hybridize to form a protein binding sequence (see figure 1). The antisense sequence may be downstream (3 ') or upstream (5') of the protein-binding sequence (or flanking in the case where more than one antisense sequence is present in the resulting hybrid) (see FIG. 1 a).

In one embodiment, the protein binding sequence is CGG-N₁₁-CCG, wherein "N" is any nucleotide comprising a purine or pyrimidine base. Suitably, the protein binding sequence is 5 'CGGCTGCTCTGATGCCG 3' (SEQ ID No: 19).

Advantageously, the antisense oligonucleotide composition comprises a pair of oligonucleotides that hybridize to form a protein binding sequence, including sequence 5 'CGGCATCAGAGCAGCCG 3' (SEQ ID No: 20). Examples of such sequences are

SEQ ID Nos

9,11 and 13 shown below (example 4).

In one embodiment, the oligonucleotide further comprises flanking sequences between the protein binding sequence and the antisense sequence (see figure 1 a).

Advantageously, the oligonucleotides used in the present invention may be sulphurised, thereby increasing their stability. In a sulphurised oligonucleotide, one or more non-bridging oxygen atoms in the oligonucleotide chain may be replaced by sulphur atoms. Sulfurization can be carried out, for example, by using Beaucage's reagent (3H-1, 2-benzodioxole-3-one-1, 1, -dioxide) (II)

Et al, 2005), and the like. Other ways of increasing the stability of oligonucleotides may be used in the present invention: such as those disclosed in Leumann (2002).

The invention also provides an in vitro method of down-regulating expression of a target gene in the cytosol of a cell, using an antisense oligonucleotide composition as described above to deliver the antisense oligonucleotide across the membrane of the cell.

An alternative aspect of the invention is an antisense oligonucleotide composition as described above for use in a method of down-regulating gene expression of Human Papilloma Virus (HPV) in Human cervical epithelial cells (see fig. 5 and example 7).

In a further aspect, the present invention provides the antisense oligonucleotide composition as described above as a potential therapeutic or prophylactic agent for cancer, particularly cancer caused by Human Papillomavirus (HPV) including cervical cancer and oral cancer.

Many plant and bacterial toxins (virulence factors) have evolved to deliver toxins in the form of large catalytically active protein domains to the cytosol of cells. Examples include cholera toxin (Sandvig et al, 2005), ricin (Sandvig et al, 2010), and anthrax toxin (Gaur et al, 2002). These substances are known to act in vivo, causing morbidity and mortality in humans.

The present inventors have prepared a known attenuated form of the Bacillus anthracis lethal factor (LFn) in which the wild-type domains II-IV (responsible for the catalytic virulence activity) have been removed using recombinant PCR. These domains (II-IV) are subsequently replaced by a nucleic acid binding domain such as Saccharomyces cerevisiae GAL 4. All of which are known in the art (Gaur et al, 2002). The inventors have shown that GAL4 can bind indirectly to an active antisense oligonucleotide under suitable conditions, i.e., when two suitable oligonucleotides hybridize together to form a hybrid comprising a single-stranded antisense sequence and a double-stranded (DS) protein binding sequence. The inventors have further shown that the antisense oligonucleotide can be translocated into the cytosol of cells by its interaction with another protein PA83 using the intact complex (LFn-GAL4:: ASO). It is noteworthy that the methods and materials used by the inventors to generate intact antisense oligonucleotide-protein complexes mean that no coagulants or polycationic affinity treatments (e.g., poly (L-lysine) (Gaur et al, 2002; WO97/23236) are required, which have been heretofore considered necessary in the art for DNA complex formation, but are toxic and known to destabilize membranes (Richardson et al, 1999)).

According to a preferred embodiment of the system of the present invention the single stranded antisense oligonucleotide sequence is provided 3 'and/or 5' to a Double Stranded (DS) protein binding sequence. Proteins such as LFn can be used conjugated to pore-forming proteins such as PA 83. PA83 has the ability to enhance membrane translocation of LFn. The interaction between LFn and PA83 is described elsewhere (Krantz et al, 2005), but it is described that it requires a globular transformation of the cargo during hole displacement. We have now shown that supramolecular assemblies can be used as cargo and that antisense sequences associated with the supramolecular assemblies can be translocated into the cytosol. This has not been predicted.

The inventors have also developed a system using hybrid oligonucleotides containing a double stranded LFn-GAL4 binding sequence and a 3 'and/or 5' overhanging single stranded antisense sequence, which have been shown to be no less effective at antisense oligonucleotide (ASO) concentrations above 30pMol than antisense (single stranded nucleotide) sequences without any flanking sequences. The complex developed by the present inventors can be used as an antisense oligonucleotide conjugation system to facilitate antisense oligonucleotide cytoplasmic delivery, but without covalent attachment between the oligonucleotide and the LFn-GAL4 protein. The present invention provides useful tools for in vitro studies, as well as a basis for a key part of the delivery method for antisense oligonucleotide pharmaceutical compositions and any suitable antisense therapy. In particular, the present invention may form the basis of an antisense antiviral composition for the treatment of HPV-infected cervical epithelial cells.

The dimerization of LFn-GAL4 protein in response to the presence of a double-stranded protein recognition nucleic acid sequence adjacent to an antisense sequence has been shown using low angle neutron scattering (SANS) (see example 5). Example 5 shows that upon addition of the hybrid oligonucleotide composition of the invention, the protein LFn-GAL4 forms a high molecular weight complex (the radius of gyration of the protein (Rg) extends from about 5nm to 25 nm).

To test the three component system according to the present invention, the inventors initially presented a prototype. The ability of the system to down-regulate genes was tested in human epithelial cells (Hela). LFn-GAL4 oligonucleotide complexes, when mixed with PA83, promote the down-regulation of a particular selection gene corresponding to the antisense sequence (see example 6). With respect to example 6, it illustrates the down-regulation of the protein syntaxin5 by western blotting and immunodetection (Suga et al, 2005). Gene specificity (and control of cell number) was further emphasized by normalizing target gene expression levels relative to the expression levels of "housekeeping" genes such as Derlin1, and by including a "non-sense" control (see figure 4). A 'nonsense' control is a control that includes single-stranded DNA that is not complementary to the target RNA of the expressed gene, and therefore is not expected to inhibit its expression.

In addition, antisense activity also shows expression of anti-HPV gene E7 (see example 7). HPV E7 is critical for the proliferation of the HPV viral life cycle and is the target for therapeutic intervention (Jonson et al, 2008).

The arrangement of two oligonucleotide strands that provides a double-stranded protein binding region or sequence while leaving a free single-stranded region of sequence with antisense activity provides a convenient conjugation method that can be used to bind antisense domains to agents already described in the art (e.g., LFn-GAL4 and PA 83). This facilitates the cytosolic delivery of antisense agents that can mediate the down-regulation of genes encoding potential drug targets (such as HPV E7).

Drawings

Specific embodiments of the present invention will now be described in more detail with reference to the accompanying drawings, in which:

FIG. 1, comprising FIGS. 1a to 1e, shows five different compositions according to the invention illustrating different topologies for establishing a binding site for a double-stranded protein from a pair of single-stranded antisense oligonucleotides;

FIG. 2 compares the inhibitory activity of a composition according to the invention with other antisense compositions in a cell-free assay.

FIG. 3, by using Small Angle Neutron Scattering (SANS), shows that LFn-GAL4 dimerizes to form higher molecular weight complexes in the presence of antisense compositions of the invention.

Figures 4 and 5 illustrate how the described system can down-regulate the expression of two different genes: downregulating the expression of synapsin 5 is shown in figure 4, downregulating the expression of HPV (serotype 18) Early (Early, E)7 is shown in figure 5;

FIG. 6 is the LFn-GAL4 nucleotide sequence (SEQ ID NO: 1);

FIG. 7 shows the LFn (LF domain I) protein sequence (SEQ ID No: 2);

FIG. 8 is the GAL4 (amino acids 1-147) protein sequence (SEQ ID No: 3);

FIG. 9 is the V5-LFn-GAL4-6His DNA sequence (SEQ ID No: 4);

FIG. 10 is the sequence of V5-LFn-GAL4-6His protein (SEQ ID No: 5);

FIG. 11 is the MRGS-6His-PA83DNA sequence (SEQ ID No: 6); and

FIG. 12 is the MRGS-6His-PA83 protein sequence (SEQ ID No: 7).

Detailed Description

Example 1: formation of shuttle proteins

LFn-GAL4 was enriched from a culture of E.coli (E.coli) in approximately 5mg/l yield. The DNA sequence encoding the protein LFn-GAL4 is SEQ ID No. 4 (FIG. 9). This sequence is in plasmid pET151/D (Invitrogen) (E.coli expression cassette).

The DNA sequences of LFn (LF domain 1) and GAL4 (amino acids 1-147) (known in the art-Gaur et al, 2002) were subcloned into the pET151/D bacterial expression cassette. The addition of a V5 epitope tag at the N-terminus and a6 × histidine affinity tag at the C-terminus allows for rapid immunodetection and affinity purification of the fusion protein from bacterial lysates, respectively.

Production of recombinant proteins chemically competent bacteria (E.coli BL21 × DE3(Invitrogen)) were transformed with purified plasmids (described above) and cultured overnight (10mL) in 2 × Yeast extract tryptone (2 × YT) bacterial broth containing ampicillin (200 μ g/mL) medium, after which they were sub-cultured (sub-culturing) in a large volume (1L)2 × YT also containing ampicillin at a similar concentration, followed by incubation for 3 hours in a orbital shaker set at 180rpm (37 ℃), followed by addition of isopropyl β -D-1-thiogalactoside (IPTG) to a final concentration of 500 μ M and incubation for a further 3 hours. bacterial pellets were prepared by centrifugation (6000 × g, 10min, 4 ℃) and lysis was performed using a French press (French press) set to 1500PSI²⁺

Resin (Clontech)) affinity chromatography column. LFn-GAL4 was eluted into 1mL fractions with 150mM imidazole. Fractions were analyzed for protein purity and concentration, pooled and dialyzed against phosphate buffered saline. The final protein preparations were evaluated by SDS-PAGE and subjected to Coomassie staining (to determine protein purity) and Western blot analysis (using V5 and 6 XHis specific antibodies).

Example 2: preparation of pore-forming proteins

PA83 was enriched from a culture of E.coli BL21 DE3 in about 5mg/l yield. Plasmids containing the DNA sequence known to encode PA83 were given by professor Les Bailie (university of Kadif). This plasmid contains the PA83 coding sequence subcloned into the bacterial expression vector pQE30 (Baillie et al, 2010) providing an N-terminal 6 × histidine affinity tag. Chemically competent bacteria were transformed with purified plasmid and cultured overnight as before. During the growth phase, 1l of the bacterial culture was incubated for 3 hours in an orbital shaker set at 180rpm (37 ℃). IPTG was then added to a final concentration of 500. mu.M and incubated for an additional 2 hours. Bacterial pellets were prepared by centrifugation (6000 Xg, 10min) and lysed as before using a French press. The bacterial lysate was further centrifuged (20000 Xg, 20 min). The resulting supernatant was passed through a6 × histidine affinity column. PA was eluted in 1mL fractions with 150mM imidazole. Protein fractions were analyzed for purity and concentration, pooled and dialyzed against Phosphate Buffered Saline (PBS). The final protein preparations were evaluated by SDS-PAGE and subjected to Coomassie staining (to determine purity) and Western blot analysis (using PA-specific antibodies).

The amino acid sequence of PA83 is shown in SEQ ID No. 7 (FIG. 12-MRGS-6His-PA83 protein).

Example 3: formation of antisense oligonucleotides containing a binding region for a double-stranded protein

Two complementary oligonucleotides, each encoding one strand of the GAL4 recognition sequence, are annealed to form a double-stranded (GAL4) protein binding sequence flanked by single-stranded antisense sequences. The ASO compositions described (SEQ ID Nos:8& 9: see example 4) were used in this example. The resulting complex is shown in FIG. 1 a. Hybridization of the antisense oligonucleotides was performed by repeated (× 10) thermal cycling, melting (1min., 94 ℃) and re-annealing (1min., 55 ℃) two partially overlapping oligonucleotides, keeping the single-stranded antisense oligonucleotide sequence free and interacting with the mRNA in a sequence-specific manner.

FIGS. 1a-1e show several possible topologies of protein binding (DS) sequences of DNA (or DNA analogs) related to antisense nucleic acid sequence(s) using the protein:: oligonucleotide conjugation strategy disclosed herein. FIG. 1a shows the two sequences

SEQ ID Nos

8 and 9, which are oriented in opposite sense and are joined together by complementary regions-bases 23 to 48 in SEQ ID No 8 bind to bases 48 to 23 of SEQ ID No 9 (SEQ ID No 9 is shown in FIG. 1a as 3 '-5'). The ASO hybrids of the present invention may or may not include other DNA sequences flanking the antisense sequence(s). As shown in FIGS. 1a-1e, the double-stranded (protein-binding) region of an ASO hybrid can be formed in a number of different ways. For example, the double-stranded region may be located between two antisense sequences (which may be the same or different) as shown in FIG. 1 a. Alternatively, the double-stranded region may be flanked by a single antisense oligonucleotide as shown in FIGS. 1b-1 e.

Example 4: preparation of DNA sequences for inhibiting the expression of specific proteins

Syntaxin 5-specific antisense oligonucleotide sequences comprising flanking and GAL 4-binding sequences: forward oligonucleotide (SEQ ID No:8)

5’AATTTGTTTGTTGAGGCTAATGCATGCCGGCTGCTCTGATGCCGGCAT 3’

Reverse oligonucleotide (SEQ ID No:9)

5’AATTTGTTTGTTGAGGCTAATGCATGCCGGCATCAGAGCAGCCGGCAT 3’

HPV (serotype 18) early (E)7 mRNA transcript-specific antisense oligonucleotide sequences comprising flanking and GAL 4-binding sequences:

forward oligonucleotide (SEQ ID No:10)

5’GGTCGTCTGCTGAGCTTTCTATGCATGCCGGCTGCTCTGATGCCGGCAT 3’

Reverse oligonucleotide (SEQ ID No:11)

5’GGTCGTCTGCTGAGCTTTCTATGCATGCCGGCATCAGAGCAGCCGGCAT 3’

TurboGFP mRNA transcript-specific antisense oligonucleotide sequences comprising flanking and GAL 4-binding sequences:

forward oligonucleotide (SEQ ID No:12)

5’GGTGCTCTTCATCTTGTTGGTATGCATGCCGGCTGCTCTGATGCCGGCAT 3’

Reverse oligonucleotide (SEQ ID No:13)

5’GGTGCTCTTCATCTTGTTGGTATGCCGGCATCAGAGCAGCCGGCATGCAT 3’

The above oligonucleotides (SEQ ID Nos:8-13) were synthesized with phosphorothioate modifications to improve the life of the antisense oligonucleotides.

The forward and reverse primer pairs (

SEQ ID Nos

8 and 9; 10 and 11; 12 and 13) were hybridized using a PCR cycler under the following conditions: heating to 94 ℃ for 1min, cooling to 55 ℃ for 1min, x 10. This annealing process produces hybrid single-and double-stranded DNA sequences according to the invention.

Example 5: formation of antisense oligonucleotide-shuttle protein (LFn-GAL4) complexes

Annealed ASO hybrids were prepared (as in example 4). These ASO hybrids were incubated with LFn-GAL4 in the appropriate stoichiometry (i.e., at a protein to oligonucleotide molar ratio of about 3: 1) in Phosphate Buffered Saline (PBS) for 30min at room temperature.

One such product so obtained was studied by Small Angle Neutron Scattering (SANS). The resulting profile is compared in FIG. 3 with the profile obtained in the absence of any antisense oligonucleotides. FIG. 3 shows the dimerization of LFn-GAL4 as determined by SANS in response to the addition of the antisense oligonucleotide composition shown in FIG. 1 a-small angle neutron scattering by LFn-GAL4 and LFn-GAL4 in PBS-DS-antisense oligonucleotides (about 0.5mg/ml of LFn-GAL4 in each example). The x-axis of FIG. 3 is in units of q/l/angstroms and the y-axis is in units of 1/cm. The upper black bold line shows the results in the presence of ASO, compared to the lower thin line in the absence of ASO. 0.01 to

The difference in LFn-GAL4 was attributed to dimerization, the reason for this dimerization was related to the formation of a part of the oligonucleotide double-stranded protein binding sequence (FIG. 1 a). In this example, a 2mm cuvette (Hellman analysis, Essex, UK) was loaded with LFn-GAL4 and LFn-GAL4:: oligonucleotide at a stoichiometric ratio of 3:1, respectively.

The intensity of this scattered radiation is a reflection of the size and shape and composition of the scatterers (i.e., LFn-GAL 4). Both individual proteins and their blends were tested. For the individual components, in principle no scattering was observed, indicating that their size was relatively small. In the mixture of LFn-GAL4 and hybridized oligonucleotide, scattering was observed, demonstrating that interaction occurred to form larger structures (i.e., LFn-GAL4:: ASO complex).

Example 6: down-regulation of target Gene expression (syntaxin 5)

One of the complexes prepared in example 5 was used to target the expression of the synapsin 5 gene in Hela cells. Pore-forming protein PA83 was used to help mediate membrane translocation of antisense oligonucleotides via interaction with LFn-GAL 4. The ASO portion of the complex consists of two annealed (as in example 4) ASO sequences (SEQ ID Nos:8& 9). Hela cells produce synaptotagmin 5 when cultured under conventional conditions. Hela cells were treated in various ways in the range of ASO complex concentrations of 0 (control), 1, 10, 50, 100 and 200 pMol/liter.

In fig. 4: the black solid bars represent Hela cells treated with PA83, LFn-GAL4, and "non-synapsin 5" (GFP-specific) ASO (formed from hybridized SEQ ID Nos:12 and 13). Open bars indicate Hela cells treated with PBS only. Both treatments showed sequence specificity for synaptic fusion protein 5 down-regulation (since the anti-GFP ASO sequence was inactive for synaptic fusion protein 5 here) and served as a negative control. The bars filled with transverse lines represent Hela cells electroporated with a hybrid ASO specific for syntaxin5 (formed by SEQ ID Nos:8 and 9). The bars filled by vertical lines represent Hela cells treated with hybrid ASO (formed by SEQ ID Nos:8 and 9) specific for PA83, LFn-GAL4 and syntaxin 5. The diagonal-packed columns represent Hela cells treated with syntaxin 5-specific hybrid ASO (formed by SEQ ID Nos:8 and 9) mixed with Oligofectamine (Invitrogen) following the manufacturer's recommendations.

The expression level of synapsin 5 in the cells was measured after 24 hours using immunoblotting. Results are expressed as% expression (y-axis) of control (Hela cells treated with PBS only). The data show that the composition according to the invention down-regulates the expression of syntaxin5 in a gene-specific manner. Shown at population level by western blot and immunodetection. The graph in FIG. 4 shows the expression level of synaptotagmin 5 and shows the effect of adding antisense oligonucleotide hybrids (3.2. mu.g/mL) (SEQ ID NOS:8 and 9) to LFn-GAL4 (50. mu.g/mL) incubated with PA (50. mu.g/mL) on Hela cells relative to the expression of the housekeeping gene Derlin1 (used to control changes in cell number). The data shown are from 3 separate experiments and demonstrate that the addition of the composition can mediate the knockdown of a specific gene antisense oligonucleotide encoding an antisense sequence (i.e., syntaxin 5). Data represent triplicates of each experiment and error bars represent standard error of the mean.

Example 7 Down-Regulation of target Gene expression (HPV18E7)

In addition to the use of different oligonucleotide sequences, experiments were conducted as described in example 6 above to examine the efficacy of the anti-HPVE 7 antisense oligonucleotide in the expression of anti-human papillomavirus (serotype 18) early stage 7 in Hela cells. In this example, the syntaxin 5-specific ASO hybrid of example 6 (formed by SEQ ID No:8 and SEQ ID No:9) was replaced by the HPV-E7-specific ASO hybrid (formed by SEQ ID No:10 and SEQ ID No:11) which also comprised a hybrid-binding region and small flanking regions. In this example, 200pM oligonucleotide was used in combination with LFn-GAL4 and PA83 (which are black bars on the graph in FIG. 5). The results are shown in FIG. 5: the% gene expression is shown on the y-axis. Control treatment i.e. PBS was taken as 100%: which is a white bar on the graph in fig. 5. Data represent triplicates of each experiment and error bars represent standard error of the mean. There was a statistical difference between the two treatments (p ═ 0.0104, measured according to the single tail, unpaired t test).

Example 8: in vitro toxicity of the compositions of the invention compared to Poly (ethylenimine)

The ASO hybrid-shuttle complex prepared as in example 5 was tested for toxicity relative to poly (ethylenimine) (PEI) (table 1). These data show the difference in toxicity of LFn-GAL4 protein after hybridization with ASO hybrids in the presence of 50 μ g/mL PA for more than 72h relative to cells exposed to different concentrations of PEI, a cationic polymer that has been extensively characterized in the literature as being capable of mediating transfection. Cell viability was normalized to untreated cells and determined by addition of MTT (Richardson et al, 1999).

TABLE 1

	IC in HeLa₅₀(microgram/ml)	IC in Vero₅₀(microgram/ml)
			25kDa branched PEI	2.9+/-0.6	7.3+/-0.1
0.8kDa branched PEI	2.4+/-0.2	7.4+/-0.3
			20kDa Linear PEI	3.0+/-0.1	6.9+/-0.5
PA83	>100	>100
			ASO	>100	>100
GAL4-LFn	>100	>100
			ASO::GAL4-LFn+PA83	>100	>100

Example 9 antisense inhibition of in vitro translation assays Using different antisense oligonucleotide compositions

Figure 2 shows the effect of the composition of the invention compared to component ASO in a cell-free assay (where membrane translocation is not rate-limiting).

The control reaction was performed using a "One-step human high yield mini-in vitro translation kit" (Thermo Scientific), together with a control plasmid encoding the protein turboGFP (green fluorescent protein-Evrogen). The reaction was performed in a volume of 4.5. mu.l (incubation at 30 ℃ for 3h) and turboGFP expression was monitored by Western blotting using anti-6 His primary antibody (1:1000 dilution) and anti-murine HRP-conjugated secondary antibody (1: 1000). Detection was performed with Picostatic ECL kit (GE Healthcare). At the beginning of the aforementioned 3h incubation period, three anti-turbo GFP antisense oligonucleotide (ASO) compositions were added to the reaction at different concentrations.

The results are shown in figure 2, (n ═ 3 ± SEM), relative to controls without antisense oligonucleotides. In FIG. 2, the y-axis represents the expression (%) of turboGPF relative to untreated controls, while the x-axis represents the concentration of added reacted antisense oligonucleotide (in pMol). The data points, indicated in squares, are from the single-stranded ASO oligonucleotide SEQ ID No:12, consisting of (one strand of) the turboGPF antisense sequence and the protein-binding sequence. The data points represented by triangles were generated only from the antisense sequence (i.e., 21-monomer unit (mer) GGTGCTCTTCATCTTGTTGGA, SEQ ID No:21) incorporated into SEQ ID No: 12. Data points shown as circles were generated from the ASO hybrids of the present invention formed by annealing together SEQ ID Nos 12 and 13. The results shown in figure 2 demonstrate that the presence of protein binding sequences in ASO hybrids does not inhibit the activity of antisense sequences at ASO concentrations above 30pMol (in vitro).

The invention can be used to prepare antisense compositions in which the antisense activity (as determined in a cell-free assay) of the antisense composition is not substantially reduced relative to an antisense control that does not have a protein-binding sequence or flanking sequence at antisense oligonucleotide (ASO) concentrations of no less than 30 pMol. This is illustrated by example 9 (fig. 2) above.

Reference documents:

Baillie L.W.(2009)Is new always better than old？:The development ofhuman vaccines for anthrax.Hum Vaccine,5(12):806–16.

Baillie L.W.,Huwar T.B.,Moore S.,Mellado-Sanchez G.,Rodriguez L.,Neeson B.N.,Flick-Smith H.C.,Jenner D.C.,Atkins H.S.,Ingram R.J.,AltmannD.M.,Nataro J.P.,and Pasetti M.F.(2010)An anthrax subunit vaccine candidatebased on protective regions of Bacillus anthracis protective antigen andlethal factor.Vaccine,28(41),pp.6740–6748.

Biroccio A.,Leonetti C.,and Zupi G.(2003)The future of antisensetherapy:combination of anticancer treatments.Oncogene,22:6579-6588.

Campbell R.N.,Leverentz M.K.,Ryan L.A.,Reece R.J.(2008).Metaboliccontrol of transcription:Paradigms and lessons from Saccharomycescerevsiae.Biochem J.414(2):177-87.

J.,Ausín C.,Chmielewski M.K.,Kauffman J.S.,Snyder J.,Del-Grosso A.,Beaucage S.L.(2005)NMR study of the desulfurization ofoligonucleoside phosphorothioates effected by"aged"trichloroacetic acidsolutions.J.Org.Chem.70(8):3303-6.

Gaur R.,Gupta P.K.,Goyal A.,Wiles W.,and Singh Y.(2002)Delivery ofnucleic acid into mammalian cells by anthrax toxin.Biochemical andBiophysical Research Communications.297:1121-1127.

Jonson A.L.,Rogers L.M.,Ramakrishnan S.and Downs Jr L.S.(2008),Genesilencing with siRNA targeting E6/E7 as a therapeutic intervention in a mousemodel of cervical cancer.Gynecologic Oncology.111(2):356–364.

Kim S.H.,Mok H.,Jeong J.H.,Kim S.W.,Park T.G.(2006)Comparativeevaluation of target-specific GFP gene silencing efficiencies for antisenseODN,synthetic siRNA,and siRNA plasmid complexed with PEI-PEG-FOLconjugate.Bioconjug Chem.17(1):241-4.

Krantz B.A.,Melnyk R.A.,Zhang S.,Juris S.J.,Lacy D.B.,Wu Z.,Finkelstein A.,Collier R.J..(2005)A phenylalanine clamp catalyzes proteintranslocation through the anthrax toxin pore Science.309(5735):777-81.

Kushner N.,Zhang D.,Touzjian N.,Essex M.,Lieberman J.,and Lu Y.(2003)A Fragment of Anthrax Lethal Factor Delivers Proteins to the Cytosol withoutRequiring Protective Antigen.Proc Natl Acad Sci U S A,100(11):6652-6657.

Leumann C.,DNA analogues:from supramolecular principles to biologicalproperties.(2002)Bioorg. Med Chem.；10(4):841-54.

Lysik M.A.,Wu-Pong S.(2003)Innovations in oligonucleotide drugdelivery.J.Pharm.Sci.92:1559-1573.

Parkin,DM.(2006)The global health burden of infection-associatedcancers in the year 2002.Int.J.Cancer.118:3030-3044.

Richardson S.,Ferruti P.and Duncan R.(1999)Poly(amidoamine)s aspotential endosomolytic polymers:Evaluation in vitro and body distribution innormal and tumour bearing animals.J.Drug Targeting,6(6),391-404.

Roehr B,(1998)Fomivirsen approved for CMV retinitis.J.Int.PhysiciansAIDS Care.4(10):14-6.

Sandvig K.,and van Deurs B.(2005).Delivery into cells:Lessons learnedfrom plant and bacterial toxins.Gene Therapy.(12)s 865-872.

Sandvig,Kirsten；Torgersen,Maria Lyngaas；Engedal,Kim Nikolai；Skotland,Tore&Iversen,Tore Geir(2010).Protein toxins from plants and bacteria:Probesfor intracellular transport and tools in medicine.FEBS Letters.584(12)s 2626-2634.

Suga K.Hattori H.and Saito A.(2005).RNA interference-mediatedsilencing of the syntaxin 5 gene induces Golgi fragmentation but capable oftransporting vesicles.FEBS Letters.579(20):4226–4234.

US Patent WO97/23236:The use of toxin peptides and/or affinityhandles for delivering compounds into cells.

Yacyshyn B.R.,Bowen-Yacyshyn M.B.,Jewell L.,Rothlein R.A.,MainolfiE.,Tami J.A.,Bennett C.F.,Kisner D.L.,and Shanahan W.R.(1998)A placebo-controlled trial of ICAM-1 antisense oligonucleotide in the treatment ofsteroid-dependent Crohn’s disease.Gastroenterology.114:1133-1142.

Zornetta I.,Brandi L.,Janowiak B.,Dal Molin F.,Tonello F.,CollierJ.and Montecucco C.(2010)Imaging the cell entry of the anthrax oedema andlethal toxins with fluorescent protein chimeras,Cellular Microbiology,12(10)；1435-45.

Claims

1. an antisense oligonucleotide composition useful for delivering an antisense oligonucleotide into the cytosol of a cell, the composition comprising:

a pair of partially complementary single stranded oligonucleotides, wherein the two oligonucleotides hybridize to give at least one single stranded antisense sequence and at least one double stranded protein binding sequence;

and a shuttle protein capable of transporting genetic material through the pores of a cell membrane, the shuttle protein comprising a nucleic acid binding domain that recognizes and non-covalently binds to a double-stranded protein binding sequence; wherein the shuttle protein comprises the Bacillus anthracis lethal factor Domain I; and

pore-forming protein, which is protective antigen PA83 of bacillus anthracis virulence factor,

the single stranded antisense sequence flanks a partially overlapping second oligonucleotide strand that forms a binding site that can bind to a nucleic acid binding domain of a protein.

2. The composition of claim 1, comprising an oligonucleotide modified by sulfurization to increase stability.

3. The antisense oligonucleotide composition according to any of the preceding claims, wherein the protein binding sequence is CGG-N₁₁-CCG, wherein "N" is any purine or pyrimidine base.

4. The antisense oligonucleotide composition according to claim 1 or 2, wherein the protein binding sequence is:

5'-CGGCTGCTCTGATGCCG-3' or 5'-CGGCATCAGAGCAGCCG-3'.

5. The antisense oligonucleotide composition of claim 1 or 2, wherein at least one of the domains of the Bacillus anthracis lethal factor is replaced with the nucleic acid binding domain.

6. The antisense oligonucleotide composition of claim 1 or 2, wherein the nucleic acid binding domain of the shuttle protein is Saccharomyces cerevisiae GAL 4.

7. The antisense oligonucleotide composition according to claim 1, wherein the antisense sequence is a single-stranded DNA designed to hybridize to messenger RNA derived from a target gene.

8. The antisense oligonucleotide composition of claim 7 wherein the target gene is expressed by a virus.

9. The antisense oligonucleotide composition of claim 7 or 8, wherein the target gene is expressed by human papilloma virus.

10. An in vitro method of down-regulating expression of a target gene in the cytosol of a cell, the method comprising delivering the antisense oligonucleotide composition of any one of claims 1-9 and across the membrane of the cell.

11. The antisense oligonucleotide composition according to any one of claims 1,2, 7 or 8, for use in medical treatment, and optionally further comprising at least one additive selected from the group consisting of pharmaceutically acceptable excipients and carriers.

12. The antisense oligonucleotide composition of claim 11, for treating a patient infected with human papillomavirus.

13. The antisense oligonucleotide composition according to claim 11, for use in the treatment of cancer.

14. A method of non-covalently conjugating an antisense oligonucleotide to a shuttle protein, the method comprising:

(i) providing two partially complementary single stranded oligonucleotides,

(ii) hybridizing two oligonucleotides to form an oligonucleotide hybrid, wherein the hybrid comprises at least one single-stranded antisense sequence and at least one double-stranded protein binding sequence,

(iii) providing a shuttle protein capable of transporting genetic material through the pores of a cell membrane and having a nucleic acid binding domain recognizing the binding sequence of the double-stranded protein, and providing a pore-forming protein comprising the Bacillus anthracis lethal factor domain I, said pore-forming protein being the protective antigen PA83 of the Bacillus anthracis virulence factor, and

(iv) non-covalently conjugating the shuttle protein to the antisense hybrid with the aid of the nucleic acid binding domain of the shuttle protein and the protein binding sequence of the oligonucleotide hybrid,

15. The antisense oligonucleotide composition of claim 3, wherein the protein binding sequence is:

5'-CGGCTGCTCTGATGCCG-3' or 5'-CGGCATCAGAGCAGCCG-3'.

16. The antisense oligonucleotide composition of claim 3 wherein at least one of the domains of the Bacillus anthracis lethal factor is replaced with the nucleic acid binding domain.

17. The antisense oligonucleotide composition of claim 4, wherein at least one of the domains of the Bacillus anthracis lethal factor is replaced with the nucleic acid binding domain.

18. The antisense oligonucleotide composition of claim 15 wherein at least one of the domains of the bacillus anthracis lethal factor is replaced with the nucleic acid binding domain.

19. The antisense oligonucleotide composition of claim 3 wherein the nucleic acid binding domain of the shuttle protein is Saccharomyces cerevisiae GAL 4.

20. The antisense oligonucleotide composition of claim 4 wherein the nucleic acid binding domain of the shuttle protein is Saccharomyces cerevisiae GAL 4.

21. The antisense oligonucleotide composition of claim 5 wherein the nucleic acid binding domain of the shuttle protein is Saccharomyces cerevisiae GAL 4.

22. The antisense oligonucleotide composition of any one of claims 15-18, wherein the nucleic acid binding domain of the shuttle protein is saccharomyces cerevisiae GAL 4.

23. The antisense oligonucleotide composition according to claim 2, wherein the antisense sequence is a single-stranded DNA designed to hybridize to messenger RNA derived from a target gene.

24. The antisense oligonucleotide composition according to claim 3, wherein the antisense sequence is a single-stranded DNA designed to hybridize to messenger RNA derived from a target gene.

25. The antisense oligonucleotide composition according to claim 4, wherein the antisense sequence is a single-stranded DNA designed to hybridize to messenger RNA derived from a target gene.

26. The antisense oligonucleotide composition according to claim 5, wherein the antisense sequence is a single-stranded DNA designed to hybridize to messenger RNA derived from a target gene.

27. The antisense oligonucleotide composition according to claim 6, wherein the antisense sequence is a single-stranded DNA designed to hybridize to messenger RNA derived from a target gene.

28. The antisense oligonucleotide composition according to any of claims 15-21, wherein the antisense sequence is a single-stranded DNA designed to hybridize to messenger RNA derived from a target gene.

29. The antisense oligonucleotide composition according to claim 22, wherein the antisense sequence is a single-stranded DNA designed to hybridize to messenger RNA derived from a target gene.

30. The antisense oligonucleotide composition of any of claims 23-27, 29 wherein the target gene is expressed by a virus.

31. The antisense oligonucleotide composition of claim 28 wherein the target gene is expressed by a virus.

32. The antisense oligonucleotide composition of any of claims 23-27, 29, 31 wherein the target gene is expressed by human papillomavirus.

33. The antisense oligonucleotide composition of claim 28 wherein the target gene is expressed by human papilloma virus.

34. The antisense oligonucleotide composition of claim 30 wherein the target gene is expressed by human papilloma virus.

35. An in vitro method of down-regulating expression of a target gene in the cytosol of a cell, the method comprising delivering the antisense oligonucleotide composition of any one of claims 15-34 and across the membrane of the cell.

36. The antisense oligonucleotide composition according to claim 3, for use in medical treatment and optionally further comprising at least one additive selected from the group consisting of pharmaceutically acceptable excipients and carriers.

37. The antisense oligonucleotide composition according to claim 4, for use in medical treatment and optionally further comprising at least one additive selected from the group consisting of pharmaceutically acceptable excipients and carriers.

38. The antisense oligonucleotide composition according to claim 5, for use in medical treatment and optionally further comprising at least one additive selected from the group consisting of pharmaceutically acceptable excipients and carriers.

39. The antisense oligonucleotide composition according to claim 6, for use in medical treatment and optionally further comprising at least one additive selected from the group consisting of pharmaceutically acceptable excipients and carriers.

40. The antisense oligonucleotide composition according to claim 9, for use in medical treatment and optionally further comprising at least one additive selected from the group consisting of pharmaceutically acceptable excipients and carriers.

41. An antisense oligonucleotide composition according to any of claims 15-21, 23-27, 29, 31, 33-34 for use in medical treatment and optionally further comprising at least one additive selected from the group consisting of pharmaceutically acceptable excipients and carriers.

42. The antisense oligonucleotide composition according to claim 22, for use in medical treatment and optionally further comprising at least one additive selected from the group consisting of pharmaceutically acceptable excipients and carriers.

43. The antisense oligonucleotide composition according to claim 28, for use in medical treatment and optionally further comprising at least one additive selected from the group consisting of pharmaceutically acceptable excipients and carriers.

44. The antisense oligonucleotide composition according to claim 30, for use in medical treatment and optionally further comprising at least one additive selected from the group consisting of pharmaceutically acceptable excipients and carriers.

45. The antisense oligonucleotide composition according to claim 32, for use in medical treatment and optionally further comprising at least one additive selected from the group consisting of pharmaceutically acceptable excipients and carriers.

46. The antisense oligonucleotide composition of any one of claims 15-21, 23-27, 29, 31, 33-34, 36-40, 42-45 for use in treating a patient infected with human papillomavirus.

47. The antisense oligonucleotide composition of claim 22, for treating a patient infected with human papillomavirus.

48. The antisense oligonucleotide composition of claim 28, for treating a patient infected with human papillomavirus.

49. The antisense oligonucleotide composition of claim 30, for treating a patient infected with human papillomavirus.

50. The antisense oligonucleotide composition of claim 32, for treating a patient infected with human papillomavirus.

51. The antisense oligonucleotide composition of claim 41, for treating a patient infected with human papillomavirus.

52. An antisense oligonucleotide composition according to any one of claims 15-21, 23-27, 29, 31, 33-34, 36-40, 42-45 for use in the treatment of cancer.

53. The antisense oligonucleotide composition according to claim 22, for use in the treatment of cancer.

54. The antisense oligonucleotide composition according to claim 28, for use in the treatment of cancer.

55. The antisense oligonucleotide composition according to claim 30 for use in the treatment of cancer.

56. The antisense oligonucleotide composition according to claim 32, for use in the treatment of cancer.

57. The antisense oligonucleotide composition according to claim 41, for use in the treatment of cancer.

58. The antisense oligonucleotide composition of any one of claims 13, 53-57, wherein the cancer is cervical cancer or oral cancer.

59. The antisense oligonucleotide composition of claim 52, wherein the cancer is cervical cancer or oral cancer.